I’m late to this thread and it’s only tangently related, but I’ve got to wonder if the mentioned-by-Linus secondary removable battery that can go into the dGPU/m.2 expansion port can/will have its own type-C connector so that:
you can charge the battery even while its removed from the laptop (that way you can have two and charge one battery while using the other battery or the like)
free up one of the USB-C expansion slots if you really want to maximize the amount of I/O
People in other threads have done deep dives into the connector for the expansion bay. It very much does not seem to be designed for hot swapping. As in USB C is rated for 10000 insertion/removal cycles, PCIe is designed for maybe 50-100, and this connector for even less. This is fragile, it appears to require a number of screws and removal of the keyboard tray to detach. Again, somewhat speculation but these were users diving into the schematics of the connector and bay. Don’t get too excited about hot-swapping batteries just yet. Wait until they talk about the expansion bay officially.
Deep dives? How? It’s not even a purchase-able product yet… Maybe my definition of a “deep dive” is more akin to physical disassembly than your definition?
Regardless, it still can make sense to have a type-c port directly on such secondary battery. I mean, I think Framework even mentioned the dGPUs being able to have their own video outputs or the like, and some existing desktop GPUs (the most recent being the 7900XT/X) have type-C ports on them.
I could argue though that just two screws isn’t exactly enough in my book to make it not be considered “hot swappable”
I mean, serial, parallel, VGA, and DVI connectors of old had two (admittedly thumb) screws on them as standard (though some laptops, like a Asus EeePC 1000H netbook I own, lacks the screw holes for its VGA output).
Absolutely it makes sense. I think extra ports are a great idea. I’m just cautioning against considering hot-swap as an option until FW confirms it, since the evidence indicates it very much is not. Hope for the best, but expect the worst. Pessimistic maybe, but you’ll be less disappointed.
One important aspect then would be the ability to show a removable battery’s charge level even when it’s not connected to the laptop (I have an HP dv6100 which predates internal batteries which has a button and 4 LEDs to show the rought battery percentage)
This would be all the more important if it’s not intended to be hot-swapped like you could with those laptops of old since, otherwise, the only way to be able to check its charge level would be to in fact plug it into the laptop - not an ideal solution if it were to support external charging.
Also, considering that Framework includes function in the BIOS to limit the charging level to a less-than-100% value, I’ve got to wonder if such a function would be able to be built right into the battery itself so that it can do that even when externally charged (though I wouldn’t be surprised if you’d still have to configure it via the laptop’s BIOS or the like while it’s plugged into the laptop, but the idea is that it’d sort of “remember” that even if you remove it and charge it externally)
Not quite sure I’m understanding you there. If it turns out that it is hot-swappable (awesome if that’s the case) then a charge level makes complete sense much like on a battery bank. But if there’s no hotswap then you could just check the charge level in your OS since it’s really just functionally making your internal battery bigger - there’d be no need for an external indicator. Also either way checking battery health/lifespan separate to internal battery makes a lot of sense but the expansion connector should be able to manage that and I’m sure they’ve already thought of that.
Edit: I think I maybe see what you’re getting at. The idea that the only way to charge the expansion battery is externally. That makes very little sense to me. The laptop should be able to charge both the internal battery and expansion battery through normal PD, and likewise charging through the expansion battery should be able to charge both as well. The expansion connector should easily be able to carry that current to both batteries without issue.
If anything, it’s not going to be ‘hot’-swappable. I would assume that any hot-swappability requires protocols for power and signal initialization and termination, and drivers (otherwise the OS is just asking, where did you go?), and also assuming the physical connector has that baked into the specification. If not, you’re asking for Framework to do a hell of a lot of R&D…which I don’t believe it’s currently capable of.
Hi, Have you considered having two usb-c port outputs from the charger, such that one could charge a laptop and another device with appropriate PD management?
Asking as someone who’s appreciated your design approach to the framework charger in keeping it compact and with a standard AC port over a foldout plug (which in my experience can break quickly)
In my opinion, that line simply means that the charger supports the USB PD standard, and thus can charge any device simultaneously.
We can hope though xD
Indeed, it’s likely single output. If you google around and lookup the datasheet for the WT6676F controller and products that use it, it’s single port output as far as I can tell.
There simply may not be PMICs available doing 180W+ for the USB PD SPD/EPR spec that also support dual or multiple port outputs yet. This said the Anker 737 battery bank supposedly supports multiple outputs and uses the WT6676F but they may have an additional controller going on to facilitate that.
Hey i am fairly late to this conversation and during my search in this post i couldnt find only a short discussion about a port for the secondary battery to charge it while not connected.
Many discrete GPUs in tower computer also utilize external 6/8/12 Power Connectors additionally to satisfy the need for power. Would it be possible to build something similar for the GPU Bay? It could have an own (only) USB-C PD connector at the back to pull power from. Would that be even technically possible?
That would satisfy the need of the power-hungry GPU with the 180W power adapter while charging the battery of the laptop with i.e. 90W power delivery through a dock.
Remember that every watt of power used is a watt of heat that needs to be cooled. Look at the size of the cooling system on desktop GPUs.
Once you start trying to get above the 240W that the Framework-16 already supports through its normal power port, I suspect adequate cooling might be more of a problem.
Thats certainly something i had not considered.
My main concern or interest was due the fact that most low and mid prized docking stations only support up to around 90W PD, which does net even remotely satisfy a power hungry GPU.
I want to use a docking station for hiding all cable while also utilising the full capabilities of the new framework.
For this it would be nice to have an dedicated PD port on the GPU bay. The total power draw could still be limited to 240W but this enables the usage of my currently docking station while still providing enough power to the GPU.
USB PD over 100W is pretty new. Higher power docking stations will be coming. Framework’s 180W power adapter is using one of the first chips available that can even do USB PD that high.
In that scenario, you would be using one cable for the docking station and a second cable plugged into your dedicated PD port on the GPU bay, yes?
If you’re already ok with using two cables, then you can do that already. One cable for the docking station and the other for power. You can use a couple right-angle USB cables, if you want to minimize cables sticking straight out.
Since our discussion the FW 16 was released, rendering this conversation a little bit useless. I have orderd mine (only got Q1/2024 ) but still very happy to be part of this community.
For the connectivity of two USB-C cable, is it possible to have a docking station with 90W PD for daily use and connecting an additional USB-C PD with more power if necessary. Will USB PD automatically switch to the higher charger or do i need to follow a specific procedure?
On the FW13 it will charge from the PD source that offers the most power, so I’d assume the same logic will be in the FW16. No procedure, it’ll just only draw from the most powerful source.